Nearly three million elderly are housed in long-term care (LTC) settings, and the numbers and proportions of persons who live and die there are increasing. Until recently, most LTC was provided in nursing homes (NHs), but non-nursing home residential care/assisted living (RC/AL) has been proliferating as an alternative to NH care. RC/AL facilities are extremely varied, ranging from small board and care homes to large complexes; in many cases, their residents resemble persons in NHs, including elderly with Alzheimer's Disease and related dementia. Despite their prevalence, and the fact that they are becoming significant providers of end-of-life care, virtually no information exists regarding how LTC facilities attend to the end-of-life of their residents; for RC/AL facilities, such information is completely lacking. This study will collect data from an established, stratified, random sample of LTC facilities and residents across four states to describe how care is provided in these diverse RC/AL facilities (stratified to include small facilities; larger, traditional facilities; and larger, new-model facilities) and NHs, and will compare the structure and process of that care, select outcomes of care, and the relationship between care and outcomes for a diverse group of residents. Specifically, data will be collected from 193 RC/AL facilities and 40 NHs participating in NIA's Collaborative Studies of Long-Term Care, to describe, compare, and evaluate the structure (the facility's capacity to provide care; the care setting) and process (the manner in which care is delivered; the application of care) of end-of-life care. Data also will be collected for a stratified sample of 450 of the residents who die or are transferred up to three days before their death, to describe: (1) the characteristics of the residents who die, such as their age, cognitive and comorbid status, and cause and site of death; (2) the care provided to these residents at the end-of-life, such as primary careprovider continuity and training to manage pain (structure), and advance care planning, administration of pain medications, and emotional support (process); and (3) select outcomes of end-of-life care, such as resident discomfort and quality of life, and family and staff satisfaction with care. Analyses also will determine the relationship between the structure and process and select outcomes of end-of-life care [e.g., the relationship between careprovider training (structure), administration of pain medications (process), and resident discomfort (outcome)]. This project will constitute a significant advance in the data available to maximize the end-of-life experience of millions of elderly who die in LTC settings.